2020-04-17 CANCELLED: Using R at LRZ (HURL1S20)

R is a highly popular and powerful programming language for data analysis and graphics, used in many research domains. The Leibniz Supercomputing Centre (LRZ) is addressing the needs of R users by facilitating various ways of working with R on LRZ systems.

For one it is hosting a RStudio Server web service that provides an easy to use and powerful platform mostly targeted at interactive analyses. This service can, however, also be used as a gateway to the high performance computing (HPC) systems at LRZ. Additionally, R can be employed directly on the massively parallel Linux Cluster and SuperMUC-NG as well as on specialized and GPU-enhanced machine learning (ML) systems.

Schedule (tentative):

• Short introduction to the LRZ (B)
• Working with selected features of RStudio Server (R notebooks, built-in terminal, database connections, session management) (B) / (I)
• R on the LRZ Linux Cluster: R environment modules, R package management (B) / (I)

15 min break

• R on the LRZ Linux Cluster: Slurm Workload Manager, interactive session, job processing (B) / (I)
• Parallelization scenarios and R resources (I)
  
• Perfectly parallel workloads and multiple R scripts (B) / (I)

15 min break

• Introduction to worker queue scenario/loose coupling (batchtools, rredis/doRedis) (I) / (A)
• Shared memory parallelization (parallel/doParallel, foreach) (I) / (A)
• Message passing (rmpi/doMPI) (I) / (A)
• Futures/Promises (parallel, future, doFuture) (A)
• Workflow management (drake) (A) / (C)

Basic knowledge of R.

Certificates of attendance for

• Introduction to GNU/Linux and SSH (or comparable previous knowledge)
• Introduction to the LRZ HPC Infrastructure (or comparable previous knowledge)

All participants are expected to bring their own laptops.